HeyJaredSurviving the 9 to 5
“Surviving the 9 to 5” isn’t about hustling harder or chasing the next big thing—it’s a raw, no-bull guide for anyone stuck in the daily grind.
- Indexed episodes, last 90 days
- 20
- Latest publication
- Sep 17, 2026
- Audience
- Checking…
- Earliest in this view
- Jul 28, 2026
Latest episodes
Fixing AI context window amnesia (opens the original)
Read excerpt
This audio episode explains the conversation retention strategy in detail—breaks it down in plain, everyday English: how it works, why it’s effective, and how to implement it step by step in real life.
The memory eater (opens the original)
Read excerpt
This video episode outlines a strategic framework for managing the context window of Large Language Models to prevent silent truncation, which occurs when an AI unknowingly "forgets" early parts of a conversation to make room for new data. The video defines working memory as a reserved portion of the prompt used for step-by-step reasoning and explains that conversation limits are reached when the combined total of input and output tokens exceeds the model’s capacity. To maintain long-term accura
Context Budget Protocol (opens the original)
Read excerpt
A video episode breaking down the four categories of the context budget (System, Docs, History, Working Memory), explaining the difference between input and output token pools, and showing how to structure a prompt systematically.
Stop Treating AI Context Like A Dumpster (opens the original)
Read excerpt
This audio episode walks you through the Context Budget Allocation hack with concrete examples showing how it works in practice—what goes in, what comes out, and why the allocation matters. which token pool do the four categories (System instructions, Retrieved documents, Conversation history, Working memory/scratchpad) get subtracted from—input tokens or output tokens? Convert 8,000 tokens into familiar units (words, pages, etc.) so you can intuit the actual volume. If I use a RAG system with 2
Decoding AI Documents (opens the original)
Read excerpt
Decoding AI Documents
Publishing over time
Last 90 days. Choose a month to open its work.
Recurring subjects
Named in the text we hold. One piece can cover several.
Not enough subject data for this period yet.
Audience
No verified audience measurement yet.
About this data
Counts cover the work we have indexed. Tone needs enough text and a confident classification. Excerpts and episode notes are not full articles or transcripts.
Identity or attribution wrong? Suggest a correction.